From 8d1686b7f63b42f8e3547ab39f5a977df072be72 Mon Sep 17 00:00:00 2001
From: Ona De Gibert Bonet <degibert@mahti-login15.mahti.csc.fi>
Date: Tue, 1 Oct 2024 10:16:28 +0300
Subject: [PATCH] added configuration for korean-english

---
 configs/hplt/config.hplt.kor-eng.yml | 39 ++++++++++++++++++++++++++++
 1 file changed, 39 insertions(+)
 create mode 100644 configs/hplt/config.hplt.kor-eng.yml

diff --git a/configs/hplt/config.hplt.kor-eng.yml b/configs/hplt/config.hplt.kor-eng.yml
new file mode 100644
index 000000000..4ff87ecbd
--- /dev/null
+++ b/configs/hplt/config.hplt.kor-eng.yml
@@ -0,0 +1,39 @@
+####
+# Example of a production config
+# Change language pair, experiment name, datasets and other settings if needed
+# Training low resource languages might require more tuning of pipeline/training/configs
+###
+
+experiment:
+  dirname: hplt
+  name: kor-eng
+  langpairs:
+    - ko-en
+
+  #URL to the OPUS-MT model to use as the teacher
+  opusmt-teacher: "https://object.pouta.csc.fi/Tatoeba-MT-models/kor-eng/opusTCv20210807-sepvoc_transformer-big_2022-07-28.zip"
+  
+  # Specify if the teacher and the student are many2one
+  one2many-teacher: False
+  one2many-student: False
+
+  #URL to the OPUS-MT model to use as the backward model
+  opusmt-backward: ""
+
+  teacher-ensemble: 1
+  
+  split-length: 20000000
+  spm-sample-size: 2000000
+
+  best-model: perplexity
+
+  opusfilter:
+    config: default
+
+datasets:
+  train:
+    - tc_Tatoeba-Challenge-v2023-09-26
+  devtest:
+    - flores_dev
+  test:
+    - flores_devtest
\ No newline at end of file